Paper2026
Material reviewed: AbstractThe abstract describes Honey Shield, a deception-driven cybersecurity paradigm for web authentication systems. It proposes a centralized security service interface offering defenses such as honeypot redirection and SQL injection protection, redirecting malicious users to a decoy system and logging incidents for later analysis.
ApplicationIdentityHoneypotDecoy
Paper2026
Material reviewed: AbstractThe work presents HoneyPrompt, a multi-sensor fusion architecture for detecting and classifying autonomous AI agents that interact with industrial control systems. The authors deploy a maritime SCADA honeypot simulating a pipeline monitoring gateway across six protocols and instrument it with a graduated prompt injection acting as a sensor array.
OT/ICSNetworkHoneypotHoneytoken
Paper2026
Material reviewed: AbstractThis paper adapts the Honeyquest instrument to automatically evaluate LLM-based attacker judgment against cyber deception techniques. According to the authors, a cohort of 21 models from various providers was compared against a human baseline on reconnaissance tasks, finding that LLMs fall for traps at higher rates than humans and show a gap between recognizing deception in their reasoning and acting accordingly. The authors conclude that human-centered deception hypotheses do not reliably transfer to AI attackers.
NetworkHoneypotDecoy
Paper2026
Material reviewed: AbstractThe authors present HoneyRoute, an inference-serving layer that detects malicious requests and routes them to a dedicated honeypot model, shielding the production model while harvesting attacker intelligence. The system combines a streaming router, a dual-implementation honeypot, and an analysis loop that feeds back into router retraining.
ApplicationCloudHoneypotDecoy
Paper2026
Material reviewed: AbstractAccording to the authors, HoneyTrap is a deceptive defense framework for large language models that employs four collaborative defender agents to counter multi-turn jailbreak attacks by directing attackers toward honeypot traps. The authors also introduce a progressive multi-turn jailbreak dataset (MTJ-Pro) and two new evaluation metrics for deceptive defense. They report reductions in attack success rates across several commercial and open-source models.
ApplicationHoneypotAdversary engagement
Paper2026
Material reviewed: AbstractThis work presents Honeyval, an evaluation framework for LLM-powered HTTP honeypots. According to the authors, the framework uses AI hacking agents as attackers and diverse backend applications to reproducibly evaluate the performance and detectability of these honeypots compared to prior limited evaluation methods.
NetworkApplicationHoneypotDecoy
Paper2026
Material reviewed: AbstractThe authors introduce ICSLure, a modular honeynet framework intended to emulate realistic industrial control system environments. It integrates physical PLCs with live data sources over industrial protocols and virtualized network components, with monitoring to log attacker interactions.
OT/ICSNetworkHoneypotAdversary engagement
Paper2026
Material reviewed: AbstractThis paper studies how to improve shell emulation accuracy of local open-weight LLM-based SSH honeypots through prompt design and supervised fine-tuning, comparing eight models against their base versions. Using 34 automated unit tests, the authors report that prompt design has a large effect and that fine-tuning effectiveness depends on dataset coverage, with an expanded dataset built from honeypot logs producing stronger local models according to their own measurements.
NetworkEndpointHoneypot
Paper2026
Material reviewed: AbstractThis study examines international differences in hacking attempts using honeypot data, analyzing 504,877 login attempts from 314 unique IP addresses over a 3-month period. The authors test hypotheses based on socio-economic factors — unemployment, poverty, GDP and education expenditures — and attribute to higher unemployment and lower education expenditures a correlation with more breach attempts per country, and to lower education expenditures a correlation with higher success rates of breach attempts.
NetworkHoneypot
Paper2026
Material reviewed: AbstractThis paper presents a noise-aware measurement study of Internet-facing DICOM services, combining IPv4-wide active scanning with passive honeypot deployments. The authors introduce a reproducible method to filter false positives caused by honeypots, network telescopes, and other measurement artifacts, reducing apparent exposure by 39%. After filtering, they identify thousands of vulnerable deployments lacking encryption and authentication, finding that existing deception systems are insufficient to capture the actual risks.
NetworkApplicationHoneypotDecoy
Paper2026
Material reviewed: AbstractAccording to the authors, this paper proposes 'Knowledge Trap', a defense against large language model extraction attacks that redirects attackers toward low-transferability knowledge using a honeypot-style knowledge graph and breadcrumb-guided exploration, rather than blocking queries or perturbing outputs. The authors report experiments in medical and financial domains showing reduced surrogate agreement without degrading legitimate-user performance.
ApplicationHoneypot
Paper2026
Material reviewed: AbstractThe study implements a deception-driven security model (D-DSM) for a campus network environment, with decoy resources managed through a centralized deception mechanism. It aims to divert malicious activity away from critical assets, provide indicators of attack behavior, and complement existing defense-in-depth controls by improving visibility of advanced and lateral attacks while helping reduce false alerts.
NetworkDecoyAdversary engagement
Paper2026
Material reviewed: AbstractThis work studies how large language model (LLM)-based penetration testing agents can allocate their execution budget when encountering honeypots. The authors formalize the attacker's problem as a budgeted decision process in which the agent must decide whether to continue exploitation or skip a target upon honeypot suspicion, proposing a detector-guided policy evaluated in a controlled mixed-host testbed.
NetworkEndpointHoneypotAdversary engagement
Paper2026
Material reviewed: AbstractThe paper proposes an LLM-enabled cloud-native dynamic honeypot architecture in which the model acts only as a text synthesizer behind policy gating and state verification. The authors describe a session broker that classifies attacker commands rather than executing them, alongside ethical governance controls and an evaluation protocol.
CloudNetworkHoneypotAdversary engagement
Paper2026
Material reviewed: AbstractThe authors propose MemPot, a defense framework that injects optimized honeypots into the memory of large language model (LLM)-based agents to counter extraction attacks. Through a two-stage optimization process and a detection model based on Wald's Sequential Probability Ratio Test, the authors report improved detection performance over existing baselines in their experiments.
ApplicationHoneypotDecoy
Paper2026
Material reviewed: AbstractThe authors present Mirage Mesh, a lightweight cyber deception framework for Linux environments that applies deception-in-depth across four layers: an SSH decoy built on Cowrie, a system-layer decoy administrative account, a fake Flask web administration interface, and a data-layer honeytoken credential file. The work proposes a shared DecoyAlertEvent schema to normalize observations from these layers, process them through a Wazuh-based detection and active-response pipeline with Python orchestration, map them to cyber kill-chain stages and MITRE ATT&CK techniques, and evaluate them with a Deception Effectiveness Metric (DEM).
NetworkEndpointHoneypotHoneytoken
Paper2026
Material reviewed: AbstractThis paper presents a context-enriched SQL injection detection framework built via a multi-agent honeypot system that generates malicious/benign requests, mediates authentic database responses while protecting production data, and pairs requests with responses to create a labeled dataset of 140,973 pairs. The authors state that models trained on this dataset outperform payload-only counterparts.
ApplicationHoneypot
Paper2026
Material reviewed: AbstractThis paper, according to its authors, reviews AI-based solutions for building network- and device-level cyber deception methods in contested environments. It focuses on combining large language models (LLMs) and reinforcement learning (RL) to optimize deception strategies such as honeypots and honeynets, and examines their validation against stealthy attacks on operational technology (OT) systems reported in the literature.
NetworkOT/ICSHoneypotDecoy
Paper2026
Material reviewed: AbstractThe authors present a cyber-risk management framework for networked digital systems combining strategic game behavior with contagion dynamics. The study addresses optimal cybersecurity resource allocation considering heterogeneous node valuations by attackers and defenders, proposing contagion-path-based risk measures and analyzing cyber deception effects through numerical experiments.
NetworkDecoy
Paper2026
Material reviewed: AbstractThe paper proposes a password-leak detection method based on Pythagorean triples, generating honeywords from a special graph-structure sequence built using number-theoretic properties of those triples. The authors describe a generation algorithm with a flat distribution and report a security proof and comparative experiments aimed at brute-force resistance and storage efficiency.
IdentityHoneytoken
Paper2026
Material reviewed: AbstractThis paper presents Ollure, a low- and medium-interaction honeypot that emulates the Ollama API without a backend LLM, as described by the authors. Deployed across four locations over 84 days, it recorded hundreds of thousands of interactions from thousands of source IP addresses. The study documents automated discovery and fingerprinting activity, as well as concrete exploitation attempts against both the infrastructure and LLM layers.
NetworkApplicationHoneypotDecoy
Paper2026
Material reviewed: AbstractThis paper introduces PHANTOM, a framework for generating honeytokens that encode organisation-specific knowledge (domain names, naming conventions, technology idioms, and secret-value distributions) through a multi-component generation pipeline. The authors propose a four-component Believability Score and use it to compare PHANTOM against a template-based baseline across several token types and organisational contexts.
NetworkApplicationHoneytoken
Paper2026
Material reviewed: AbstractThe work describes the design of a portable wireless honeypot using a Raspberry Pi to detect and analyze SSH brute-force attacks. By simulating a vulnerable SSH service with the Cowrie honeypot, the system logs login attempts and attacker behavior in real time to study attack patterns and improve security measures.
NetworkHoneypot
Paper2026
Material reviewed: AbstractThis thesis proposes game-theoretic and optimisation-based models for hardening Active Directory attack graphs. According to the authors, it introduces a honeypot/decoy placement model that minimises shortest paths and Domain Admin-reachable nodes, along with strategies accounting for the dynamic nature of the graph and adaptive prioritisation models incorporating administrator feedback. The authors note that the underlying problems are computationally intractable.
IdentityNetworkHoneypotDecoy